Are there any rules on how you rate the impact from a LV?

[MEC]

Dear all

Im working with an model which estimates which effect an commerciel has on a brand experience. What I need to do is to rate the different impacts in the model between the latent variable and the dependent variabel. In multiple regression models the rules are that if R is smaller than 0,4 its a week impact, between 0,4 - 0,8 a medium impact and over 0,8 a strong impact. Do we have any rules in the PLS on impact or is it the same rules?

Thanks

Lars

[schroer]

Hi Lars,

in general, usually the same quality criteria are used in PLS as in multiple regression regarding effect sizes, etc. You might, however, consider using f² as a more specific effect size if you're interested in the effects of single exogenous variables.

Best wishes,

Joachim

(BTW, I always wondered if there's a better way to go than to simply use the same quality criteria as in MRA. Has anybody ever tried to apply/update Cohen's (1988) chapter on canonical correlation rather than multiple regression to PLS path modelling?)